R7800 and HT160 Setting

[sfx2000]

Yes, the range will definitely help, also the fact that you're not facing a legion of non-wifi devices sharing the same band. It will still become a problem once the whole world moves to dual band routers (still a lot of single bands out there right now).

Once 30$ routers become dual band, it will start to get crowded.

Or the other elephant in the room - LTE - then 5GHz will be a bit more interesting...

 

[andy o]

I just replaced my WiFi chip to an Intel 9260 on my Razer Blade Stealth, and have the R7800. Has anyone actually achieved these two to link on 160 MHz? I have enabled HT160 on the router, and have the latest firmware (V1.0.2.44, and also didn't work in 1.0.2.40). Also installed the latest drivers 20.10.2, dated Nov 2017 from Intel's site. I see no options in Device Manager to enable this, only a HT/VHT option, which is on VHT by default. The router is displaying channels 132+136+140+144+149+153(p)+157+161, so I think it's working on that end.

The WiFi card links at 866Mbps, and there's no difference when enabling/disabling HT160 in the router with WLAN speeds, testing with LAN Speed Test and its companion server program to a desktop connected over gigabit ethernet. I'm getting about 420Mbps up / 260Mbps down less than 2 feet from the router over TCP.

 

[avtella]

HT160 worked in the SNB review when tested between two R7800 units. It’s probably Intel not enabling it yet in drivers, just a guess. I too ordered a 9260ac for testing purposes for ~$25 on eBay a week ago, hopefully it arrives by end of next week.

I find direct transfers of large files ie 1GB+ from laptop/PC to and from NAS are more accurate than LAN Speed test regardless of Packet size settings in the program. The 8265ac in my Dell Inspiron 7567 for example gives me about 68-70 MB/s (560 Mbps) Down and 40-45 MB/s (360 Mbps) Up, one floor directly below the router. My Dell Precision 7520 is showing similar results with it. The antenna placement must not be very optimal in the Razer if you are getting such low results so close to the router.

 

[andy o]

I think TCP vs UDP makes a difference. How are you testing exactly? Is UDP a reasonable way to test actually?

 

[avtella]

I did a direct transfer of a large file via Windows Explorer from and to my NAS from Laptop.

 

[_sem_]

I just replaced my WiFi chip to an Intel 9260 on my Razer Blade Stealth, and have the R7800. Has anyone actually achieved these two to link on 160 MHz? I have enabled HT160 on the router, and have the latest firmware (V1.0.2.44, and also didn't work in 1.0.2.40). Also installed the latest drivers 20.10.2, dated Nov 2017 from Intel's site. I see no options in Device Manager to enable this, only a HT/VHT option, which is on VHT by default. The router is displaying channels 132+136+140+144+149+153(p)+157+161, so I think it's working on that end.

The WiFi card links at 866Mbps, and there's no difference when enabling/disabling HT160 in the router with WLAN speeds, testing with LAN Speed Test and its companion server program to a desktop connected over gigabit ethernet. I'm getting about 420Mbps up / 260Mbps down less than 2 feet from the router over TCP.

Do you have two 160MHz bands available and clear at 5G?

It takes more than one GbE line to feed 1.73Gbps.

Check here, someone writes it works for him... http://forum.notebookreview.com/threads/intels-new-wi-fi-cards-detailed-on-the-website.802818/page-7

 

[avtella]

1733 Mbps is only theoretical, reality is usually around Gigabit speeds in terms of throughput anyway even at that link rate, in addition you have lowered range at HT160. I do have link aggregation to test with between my ReadyNAS 524X and R8500.

 

[sfx2000]

It takes more than one GbE line to feed 1.73Gbps.

Not really - WiFi is half duplex, so a wired 1Gbe link can feed it...

 

[andy o]

Do you have two 160MHz bands available and clear at 5G?

It takes more than one GbE line to feed 1.73Gbps.

Check here, someone writes it works for him... http://forum.notebookreview.com/threads/intels-new-wi-fi-cards-detailed-on-the-website.802818/page-7

I've seen that thread before, but that guy doesn't provide any details of what he's using. I do doubt that just aggregating 2 Gb ports would magically enable a 1733 Mb/s link though.

 

[avtella]

It actually doesn’t, as you said, even if somehow throughput went even a little over 1Gbps it wouldn’t really matter. 1733 is just a link rate/theoretical peak, real world speeds willl be around 930-960 Mbps which I already mentioned above plus you will also have a reduced range.

 

[avtella]

Well Dell kindly decided to replace my $700 7567 for a newer maxxed out $1,400 4K 7577, anyway after sticking in the 9260ac I am also not able to get HT160 working, maybe since I have 4 MU-MIMO (2x8265ac and 1x9260ac and a Nexus 5X) capable devices it’s prioritizing that over HT160.

 

[sfx2000]

Play around with the settings...

160MHz channels are going to have shorter range (same amount of power spread twice as wide), and MU has it's own challenges with intersymbol interference and self-noise (other STA's in the same MU frame are noise to the target STA). I did a writeup earlier this morning on some of the MU challenges that touch on the client side...

I'm thinking best performance will be 80MHz and SU mode...

 

[avtella]

I agree completely sfx2000, about 80Mhz being better, I merely wanted to test 160Mhz.

 

[BUFFBOY]

hi all,

Should this option be disabled or does it really matter ??

 

[bodean]

hi all,

Should this option be disabled or does it really matter ??

Unless you have a device that uses this (which it seems 1% of users actually do), then keep it disabled.

 

[BUFFBOY]

It's ok to leave the mu-mimo and beam forming enabled right since it's not affecting anything.

 

[Killhippie]

It's ok to leave the mu-mimo and beam forming enabled right since it's not affecting anything.

Unless you are using MU-MIMO devices I would just use SU. Beamforming may or not not help, it depends on your clients and how your network us running now, you need to have a play around and see.

 

[bodean]

It's ok to leave the mu-mimo and beam forming enabled right since it's not affecting anything.

Your best bet is just to leave the default setting on the router if this is uncharted territory for you. Change the Wireless channels to manual and select one, rest of the setting default...

 

[martopg]

I have two R7800s set up in bridge mode. Wanted to test using 160MHz channels but I don't think I've gotten it set up correctly. I have the HT160 setting checked on both routers. Looking at the info on this site and elsewhere, it seems like your primary channel has to be set to either 50 or 114 for HT160 to work, but neither of those are options on the R7800.

I tried selecting different channels (e.g. 52) and the primary router says it's doing 160 (e.g. "36+40+44+48+52(p)+56+60+64") but scanning with an analyzer app only shows half of those channels being used (e.g. 52-64).

Additionally, have not seen any notable differences between throughput with HT160 turned on vs. off.

Is there some way to confirm that these routers are actually using a 160MHz group of channels? Any other settings I need to tweak to make this work?

 

[avtella]

HT160 makes no difference if all you want is a router to router connection because in HT160 mode the R7800 acts as a dual antenna 2x2 device rather than as a 4x4 device but with double the theoretical bandwidth, so compared to the standard HT80 4x4 connection it makes no difference. Its useful more for singe/dual antenna clients like laptops and mobile devices. For router to router connection you are better off in HT80 mode.

I wouldn't necessarily trust any random monitoring app, inSSIDer detects HT160 well and so does Acyrlic so try those, plus if you client devices does not support HT160 they won't detect it as HT160 it will see it as a HT80 device so even a supporting app would make no difference in detection.